Bài tập ktl exercise
Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (143.93 KB, 14 trang )
<span class="text_page_counter">Trang 1</span><div class="page_container" data-page="1">
<b>Bài 1</b>
<b>. reg wage educ exper nonwhite female married south 519=n-k-1</b>
<b>SST=SSE+SSR bậc tự do giá trị tb của tổng bp: SS/df</b>
<b> Source | SS df MS Number of obs = 526 (n)---+--- F( 6, 519) = 41.18 SSE-Model | 2309.32588 6(k) 384.887647 Prob > F = 0.0000SSR-Residual | 4851.08841 519 9.34699115 R-squared = 0.3225---+--- Adj R-squared = 0.3147 SST-Total | 7160.41429 525 13.6388844 Root MSE = 3.0573 căn bậc 2 MS dùng để ss ………giữa các mẫu, chọn mẫu có ………..gt nhỏ hơn </b>
<b>---hs hồi quy wage | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- educ | .5694223 .0519968 10.95 0.000 .4672723 .6715723 exper | .0552699 .011024 5.01 0.000 .0336127 .076927 nonwhite | .0729731 .4437979 0.16 0.869 -.7988879 .9448342 female | -2.092037 .2717096 -7.70 0.000 -2.625823 -1.558251 married | .7150305 .2975101 2.40 0.017 .1305585 1.299503 south | -.6416218 .2830282 -2.27 0.024 -1.197644 -.0856001 _cons | -1.410051 .7743316 -1.82 0.069 -2.93126 .1111587---- hs chặn---</b>
---SRF:
wage = 1.41 + 0.569educ + 0.055exper+0.072nonwhite 2.092female+0.715married-0.641south +u^
</div><span class="text_page_counter">Trang 2</span><div class="page_container" data-page="2">Wage^ = 1.41 + 0.569educ + 0.055exper+0.072nonwhite 2.092female+0.715married-0.641south
</div><span class="text_page_counter">Trang 3</span><div class="page_container" data-page="3"><b>Bài 2</b>
<b>. reg wage educ exper nonwhite female married south, robust</b>
<b>Linear regression Number of obs = 526 F( 6, 519) = 26.91 Prob > F = 0.0000 R-squared = 0.3225 Root MSE = 3.0573--- | Robust</b>
<b> wage | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- educ | .5694223 .0657564 8.66 0.000 .4402409 .6986037 exper | .0552699 .0103954 5.32 0.000 .0348477 .075692 nonwhite | .0729731 .3861896 0.19 0.850 -.6857137 .83166 female | -2.092037 .2538627 -8.24 0.000 -2.590762 -1.593312 married | .7150305 .2672913 2.68 0.008 .1899247 1.240136 south | -.6416218 .2715594 -2.36 0.019 -1.175113 -.108131 _cons | -1.410051 .8691299 -1.62 0.105 -3.117496 .2973942---</b>
</div><span class="text_page_counter">Trang 4</span><div class="page_container" data-page="4"><b>Bài 3</b>
<b>Source | SS df MS Number of obs = 1387---+--- F( 5, 1381) = </b>
<b> Model | 28988.0176 5 5797.60351 Prob > F = 0.0000 Residual | 545486.724 1381 394.994007 R-squared = </b>
<b>---+--- Adj R-squared = 0.0470 Total | 574474.741 1386 414.48394 Root MSE = 19.874--- bwght | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- cigs | -.5014951 .0915656 -5.48 0.000 -.6811179 -.3218723</b>
<b>---cigs byte %8.0g ---cigs smked per day while pregparity byte %8.0g birth order of child</b>
<b>male byte %8.0g =1 if male childwhite byte %8.0g =1 if white</b>
<b>motheduc byte %8.0g mother's yrs of educ</b>
<b>1. Viết hàm SRF</b>
</div><span class="text_page_counter">Trang 5</span><div class="page_container" data-page="5"><b>2. Tính và giải thích ý nghĩa của hệ số xác định R</b>
<b><small>2</small></b><b>3. Kiểm định sự phù hợp của mơ hình</b>
<b>4. Kiểm định và giải thích ý nghĩa của biến cigs; male; motheduc với 3 phương pháp: kiểm định t, p-value, khoảng tin cậy</b>
<b>Bài 4</b>
<b> Source | SS df MS Number of obs = 1260---+--- F( 8, 1251) = 29.51</b>
<b> Model | 4341.73975 8 542.717468 Prob > F = 0.0000 Residual | 23005.6994 1251 18.3898477 R-squared = 0.1588</b>
<b>---+--- Adj R-squared = 0.1534</b>
<b> Total | 27347.4392 1259 21.7215561 Root MSE = 4.2883--- wage | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- looks | .6252181 .1790031 3.49 0.000 .2740387 .9763975</b>
<b>exper byte %8.0g years of workforce experiencelooks byte %8.0g from 1 to 5</b>
<b>union byte %8.0g =1 if union membergoodhlth byte %8.0g =1 if good healthblack byte %8.0g =1 if black</b>
<b>female byte %8.0g =1 if femalemarried byte %8.0g =1 if married</b>
</div><span class="text_page_counter">Trang 6</span><div class="page_container" data-page="6"><b>---+--- Adj R-squared = 0.0485 Total | 574474.741 1386 414.48394 Root MSE = 19.859--- bwght | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- faminc | .0591903 .0335575 1.76 0.078 -.006639 .1250196</b>
<b><small>bwght int %8.0g birth weight, ounces</small></b>
</div><span class="text_page_counter">Trang 7</span><div class="page_container" data-page="7"><b><small>faminc family income, $</small></b>
<b><small>cigs byte %8.0g cigs smked per day while pregparity byte %8.0g birth order of child</small></b>
<b><small>male byte %8.0g =1 if male childwhite byte %8.0g =1 if white</small></b>
<b><small>motheduc byte %8.0g mother's yrs of educ</small></b>
<b>Bài 6</b>
<b> Source | SS df MS Number of obs = 88---+--- F( 3, 84) = </b>
<b> Model | 309231.224 3 103077.075 Prob > F = 0.0000 Residual | 608623.281 84 7245.51526 R-squared = </b>
<b>---+--- Adj R-squared = 0.3132 Total | 917854.506 87 10550.0518 Root MSE = 85.121--- price | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- bdrms | 57.68736 11.49826 </b>
<b> lotsize | .0028554 .0009058 colonial | -2.202985 20.66589 _cons | 63.47836 39.90354 storage display value</b>
<b>variable name type format label variable label</b>
<b>price float %9.0g house price, $1000sbdrms byte %9.0g number of bdrms</b>
<b>---lotsize float %9.0g size of lot in square feet colonial byte %9.0g =1 if home is colonial style</b>
<b> </b>
<b>1. Write SRF</b>
</div><span class="text_page_counter">Trang 8</span><div class="page_container" data-page="8"><b>2. Calculate R-square and explain the meaning of R-squared3.Test for the overall significance of the model</b>
<b>4. Test for the significance of each independent variable with 3 methods: critical value, p-value and confidence interval; and explain the meaning of variables.</b>
<b>Intepreting the meaning of independent variablesBdrms (reject Ho)</b>
<b>- Number of bedrooms have statistically significant effect on the house price. And the </b>
effect is positive (
<i>because ^β<sub>bdrms</sub></i>>0¿.
<b>- Given the sample we have, when number of bedrooms increase by 1 room, the average price of house increases by 57.68736 thousand USD, holding other factors fixed.</b>
<b>Lotsize (reject Ho)</b>
<b>- size of lot in square feet has statistically significant effect on the house price. And </b>
the effect is positive
(<i>because ^β<small>lotsize</small></i>>0).
<b>- Given the sample we have, when size of lot increases by 1 square feet, the average price of house increases by .0028554 thousand USD, holding other factors fixed.Colonial (Not reject Ho)</b>
</div><span class="text_page_counter">Trang 9</span><div class="page_container" data-page="9"><b>- Having colonial style or not does not have statistically significant effect on the </b>
<b>---+--- Adj R-squared = 0.0782 Total | 19.4060994 140 .138614996 Root MSE = .35745--- colGPA | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- soph | .3056211 .2102452 1.45 0.148 -.1101791 .7214214</b>
<b>Campus = 1 if living in campus</b>
</div><span class="text_page_counter">Trang 10</span><div class="page_container" data-page="10"><b>Clubs = 1 if joining a club</b>
<b>Skipped: numbers of classes skipped</b>
<b>Bài 8</b>
<b> Source | SS df MS Number of obs = 173---+--- F( 2, 170) = 25.47</b>
<b> Model | 11170.8751 2 5585.43753 Prob > F = 0.0000 Residual | 37286.3735 170 219.331609 R-squared = 0.2305</b>
<b>---+--- Adj R-squared = 0.2215 Total | 48457.2486 172 281.728189 Root MSE = 14.81--- voteA | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- democA | 9.192198 2.267415 4.05 0.000 4.716283 13.66811</b>
<b> expendA | .0242475 .004022 6.03 0.000 .016308 .0321869</b>
<b> _cons | 37.87049 2.129706 17.78 0.000 33.66642 42.07457democA byte %3.2f =1 if A is democrat</b>
<b>voteA byte %5.2f percent vote for A</b>
<b>expendA float %8.2f camp. expends. by A, $1000s</b>
</div><span class="text_page_counter">Trang 11</span><div class="page_container" data-page="11">Reject Ho
<b>- Being democrat has statistically significant effect on voteA </b>
<b>- The estimated results show that being democrate will increase voteA by 9.19% </b>
compared with not being democrat, holding other factors fixed.
Reject Ho
<b>- Expenditure on compaign of A has statistically significant effect on voteA</b>
<b>- The estimated results show that when expenditure on compaign of A increase 1 unit,</b>
voteA will increase by 0.024 unit.
</div><span class="text_page_counter">Trang 12</span><div class="page_container" data-page="12"><b>Bài 9</b>
<b><small> Source | SS df MS Number of obs = 88---+--- F( 3, 84) = (SSE)Model | 309231.224 3 103077.075 Prob > F = 0.0000 (SSR)Residual | 608623.281 84 7245.51526 R-squared = </small></b>
<b><small>---+--- Adj R-squared = 0.3132 (SST)Total | 917854.506 87 10550.0518 Root MSE = 85.121--- price | Coef. Std. Err. t P>|t| [95% Conf. Interval]---+--- bdrms | 57.68736 11.49826 </small></b>
<b><small> lotsize | .0028554 .0009058 colonial | -2.202985 20.66589 _cons | 63.47836 39.90354 storage display value</small></b>
<b><small>variable name type format label variable label---price float %9.0g house price, $1000sbdrms byte %9.0g number of bdrmslotsize float %9.0g size of lot in square feetcolonial byte %9.0g =1 if home is</small></b>
</div><span class="text_page_counter">Trang 13</span><div class="page_container" data-page="13"><b>Một số ví dụ câu hỏi trắc nghiệm:</b>
them by 0.5, the new intercept and slope estimates will be:a. 4 and 2.5
b. 8 and 2.5c. 8 and 10d. 16 and 10
2. Consider the following estimated model:
^<i>Y<sub>i</sub></i>
